Adaptive AFE calibration techniques

Abstract

An adaptation engine included in a receiver of a serial data link, including at least one processor configured to: receive an error signal generated by an error slicer included in a decision feedback equalizer (DFE) and a data signal generated by a data slicer included in the DFE, perform a first iterative adaptation process corresponding to the AFE by adjusting a gain of the AFE based on the error signal and the data signal, and perform a second iterative adaptation process corresponding to the AFE by adjusting an degeneration resistance of a degeneration resistor included in the AFE based on the error signal and the data signal.

What is claimed is: 
     
       1. A receiver for a serial data link, the receiver comprising:
 an analog front end (AFE) configured to receive a serial data stream and to generate an AFE output signal based on the serial data stream; 
 a decision feedback equalizer (DFE) configured to receive the AFE output signal and to output deserialized data based on the AFE output signal, wherein the DFE comprises an error slicer configured to output an error signal based on a comparison between the AFE output signal and a reference voltage; and 
 an adaptation engine configured to perform a first adaptation process associated with the AFE, wherein the first adaptation process includes adjusting the reference voltage based on the error signal, and adjusting a gain of the AFE based on a comparison between the adjusted reference voltage and a reference voltage target. 
 
     
     
       2. The receiver of  claim 1 , wherein the DFE further comprises a data slicer configured to output a data signal based on a comparison between the AFE output signal and a zero voltage, and
 wherein to perform the first adaptation process, the adaptation engine is further configured to:
 initialize the reference voltage, 
 initialize a tap value associated with the DFE based on the reference voltage, 
 adjust the reference voltage until a value of the error signal is different from a value of the data signal to obtain the adjusted reference voltage, 
 compare the adjusted reference voltage with the reference voltage target, and 
 based on the adjusted reference voltage being less than the reference voltage target, increase the gain of the AFE. 
 
 
     
     
       3. The receiver of  claim 2 , wherein the adaptation engine is further configured to increase the gain of the AFE by adjusting a feedback resistance of at least one feedback resistor included in the AFE, and
 wherein the increasing the gain of the AFE causes an inner eye opening included in an eye diagram corresponding to the AFE output signal to be increased in a vertical direction. 
 
     
     
       4. The receiver of  claim 2 , wherein the adaptation engine is further configured to:
 based on the adjusted reference voltage being less than the reference voltage target, perform the first adaptation process again, and 
 based on the adjusted reference voltage being greater than or equal to the reference voltage target, exit the first adaptation process. 
 
     
     
       5. The receiver of  claim 2 , further comprising a phase interpolator configured to adjust a phase of a clock signal which is provided to the DFE,
 wherein the AFE comprises a degeneration resistor having a degeneration resistance, 
 wherein the adaptation engine is further configured to perform a second adaptation process associated with the AFE, and 
 wherein to perform the second adaptation process, the adaptation engine is further configured to:
 initialize the reference voltage, 
 initialize the tap value based on the reference voltage, 
 initialize the phase to a reference phase, 
 adjust the phase until the value of the error signal is different from the value of the data signal, 
 determine a current phase value by subtracting the reference phase from the adjusted phase, and 
 based on the current phase value being greater than a previous phase value, increase the degeneration resistance. 
 
 
     
     
       6. The receiver of  claim 5 , wherein the increasing the degeneration resistance causes an inner eye opening included in an eye diagram corresponding to the AFE output signal to be increased in a horizontal direction. 
     
     
       7. The receiver of  claim 5 , wherein the adaptation engine is further configured to perform the second adaptation process after the first adaptation process is performed. 
     
     
       8. The receiver of  claim 5 , wherein the adaptation engine is further configured to:
 based on the current phase value being greater than the previous phase value, perform the second adaptation process again, and 
 based on the current phase value being less than or equal to the previous phase value, exit the second adaptation process. 
 
     
     
       9. The receiver of  claim 2 , wherein to initialize the reference voltage, the adaptation engine is further configured to:
 obtain the value of the error signal and the value of the data signal, 
 based on the value of the error signal being same as the value of the data signal, increase the reference voltage, and 
 based on the value of the error signal being different from the value of the data signal, decrease the reference voltage. 
 
     
     
       10. The receiver of  claim 2 , wherein to initialize the tap value, the adaptation engine is further configured to:
 obtain the value of the error signal and the value of the data signal, 
 based on a previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being same as the value of the data signal, increase the tap value, 
 based on the previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being different from the value of the data signal, decrease the tap value, 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being different from the value of the data signal, increase the tap value, and 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being same as the value of the data signal, decrease the tap value. 
 
     
     
       11. An adaptation engine included in a receiver of a serial data link, the adaptation engine comprising:
 at least one processor configured to:
 receive an error signal generated by an error slicer included in a decision feedback equalizer (DFE), wherein the error signal is generated by the error slicer based on a comparison between a reference voltage and an analog front end (AFE) output signal, and wherein the AFE output signal is generated by an AFE based on a serial data stream, 
 receive a data signal generated by a data slicer included in the DFE, wherein the data signal is generated by the data slicer based on a comparison between a zero voltage and the AFE output signal, 
 perform a first iterative adaptation process corresponding to the AFE by adjusting a gain of the AFE based on the error signal and the data signal, and 
 perform a second iterative adaptation process corresponding to the AFE by adjusting a degeneration resistance of a degeneration resistor included in the AFE based on the error signal and the data signal. 
 
 
     
     
       12. The adaptation engine of  claim 11 , wherein to perform a current iteration of the first iterative adaptation process, the at least one processor is further configured to:
 initialize the reference voltage, 
 initialize a tap value associated with the DFE based on the reference voltage, 
 adjust the reference voltage until a value of the error signal is different from a value of the data signal, 
 compare the adjusted reference voltage with a reference voltage target, 
 based on the adjusted reference voltage being less than the reference voltage target, increase the gain of the AFE, and perform a next iteration of the first iterative adaptation process, and 
 based on the adjusted reference voltage being greater than or equal to the reference voltage target, exit the first iterative adaptation process. 
 
     
     
       13. The adaptation engine of  claim 12 , wherein the at least one processor is further configured to increase the gain of the AFE by adjusting a resistance of at least one feedback resistor included in the AFE, and
 wherein the increasing the gain of the AFE causes an inner eye opening included in an eye diagram corresponding to the AFE output signal to be increased in a vertical direction. 
 
     
     
       14. The adaptation engine of  claim 12 ,
 wherein to initialize the reference voltage, the at least one processor is further configured to:
 obtain the value of the error signal and the value of the data signal, 
 based on the value of the error signal being same as the value of the data signal, increase the reference voltage, and 
 based on the value of the error signal being different from the value of the data signal, decrease the reference voltage, and 
 
 wherein to initialize the tap value, the at least one processor is further configured to:
 obtain the value of the error signal and the value of the data signal, 
 based on a previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being same as the value of the data signal, increase the tap value, 
 based on the previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being different from the value of the data signal, decrease the tap value, 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being different from the value of the data signal, increase the tap value, and 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being same as the value of the data signal, decrease the tap value. 
 
 
     
     
       15. The adaptation engine of  claim 11 , further comprising a phase interpolator configured to adjust a phase of a clock signal which is provided to the DFE,
 wherein to perform a current iteration of the second iterative adaptation process, the at least one processor is further configured to:
 initialize the reference voltage, 
 initialize a tap value associated with the DFE based on the reference voltage, 
 initialize the phase to a reference phase, 
 adjust the phase until the value of the error signal is different from the value of the data signal, 
 determine a current phase value by subtracting the reference phase from the adjusted phase, 
 based on the current phase value being greater than a previous phase value, increase the degeneration resistance, and perform a next iteration of the second iterative adaptation process, and 
 based on the current phase value being less than or equal to the previous phase value, exit the second iterative adaptation process. 
 
 
     
     
       16. The adaptation engine of  claim 15 , wherein the increasing the degeneration resistance causes an inner eye opening included in an eye diagram corresponding to the AFE output signal to be increased in a horizontal direction. 
     
     
       17. The adaptation engine of  claim 15 ,
 wherein to initialize the reference voltage, the at least one processor is further configured to:
 obtain the value of the error signal and the value of the data signal, 
 based on the value of the error signal being same as the value of the data signal, increase the reference voltage, and 
 based on the value of the error signal being different from the value of the data signal, decrease the reference voltage, and 
 
 wherein to initialize the tap value, the at least one processor is further configured to:
 obtain the value of the error signal and the value of the data signal, 
 based on a previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being same as the value of the data signal, increase the tap value, 
 based on the previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being different from the value of the data signal, decrease the tap value, 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being different from the value of the data signal, increase the tap value, and 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being same as the value of the data signal, decrease the tap value. 
 
 
     
     
       18. A method of adapting an analog front end (AFE) included in a receiver of a serial data link, the method comprising:
 receiving an error signal generated by an error slicer included in a decision feedback equalizer (DFE) included in the receiver, wherein the error signal is generated by the error slicer based on a comparison between a reference voltage and an AFE output signal, and wherein the AFE output signal is generated by the AFE based on a serial data stream; 
 receiving a data signal generated by a data slicer included in the DFE, wherein the data signal is generated by the data slicer based on a comparison between a zero voltage and the AFE output signal; 
 performing a first iterative adaptation process corresponding to the AFE by adjusting a gain of the AFE based on the error signal and the data signal; and 
 performing a second iterative adaptation process corresponding to the AFE by adjusting a degeneration resistance of a degeneration resistor included in the AFE based on the error signal and the data signal. 
 
     
     
       19. The method of  claim 18 , wherein performing a current iteration of the first iterative adaptation process comprises:
 initializing the reference voltage; 
 initializing a tap value associated with the DFE based on the reference voltage; 
 adjusting the reference voltage until a value of the error signal is different from a value of the data signal; 
 comparing the adjusted reference voltage with a reference voltage target; 
 based on the adjusted reference voltage being less than the reference voltage target; 
 increasing the gain of the AFE, and performing a next iteration of the first iterative adaptation process; and 
 based on the adjusted reference voltage being greater than or equal to the reference voltage target, exiting the first iterative adaptation process. 
 
     
     
       20. The method of  claim 19 ,
 wherein the initializing the reference voltage comprises:
 obtaining the value of the error signal and the value of the data signal, 
 based on the value of the error signal being same as the value of the data signal, increasing the reference voltage; and 
 based on the value of the error signal being different from the value of the data signal, decreasing the reference voltage, and 
 
 wherein the initializing the tap value comprises:
 obtaining the value of the error signal and the value of the data signal; 
 based on a previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being same as the value of the data signal; increasing the tap value; 
 based on the previous value of the data signal being same as the value of the data signal, and based on the value of the error signal being different from the value of the data signal, decreasing the tap value; 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being different from the value of the data signal, increasing the tap value; and 
 based on the previous value of the data signal being different from the value of the data signal, and based on the value of the error signal being same as the value of the data signal, decreasing the tap value. 
 
 
     
     
       21. The method of  claim 18 , wherein performing a current iteration of the second iterative adaptation process comprises:
 initializing the reference voltage; 
 initializing a tap value associated with the DFE based on the reference voltage; 
 initializing a phase of a clock signal which is provided to the DFE to a reference phase; 
 adjusting the phase until the value of the error signal is different from the value of the data signal; 
 determining a current phase value by subtracting the reference phase from the adjusted phase; 
 based on the current phase value being greater than a previous phase value, increasing the degeneration resistance, and performing a next iteration of the second iterative adaptation process; and 
 based on the current phase value being less than or equal to the previous phase value, exiting the second iterative adaptation process.